Share this article on:

Measurement Properties of the Australian Women's Activity Survey


Medicine & Science in Sports & Exercise: May 2009 - Volume 41 - Issue 5 - pp 1020-1033
doi: 10.1249/MSS.0b013e31819461c2
Basic Sciences

Purpose: The Australian Women's Activity Survey (AWAS) was developed based on a systematic review and qualitative research on how to measure activity patterns of women with young children (WYC). AWAS assesses activity performed across five domains (planned activities, employment, child care, domestic responsibilities, and transport) and intensity levels (sitting, light intensity, brisk walking, moderate intensity, and vigorous intensity) in a typical week in the past month. The purpose of this study was to assess the test-retest reliability and criterion validity of the AWAS.

Methods: WYC completed the AWAS on two occasions 7 d apart (test-retest reliability protocol) and/or wore a Manufacturing Technology Inc. (MTI) ActiGraph accelerometer for 7 d in between (validity protocol). Forty WYC (mean age 35 ± 5 yr) completed the test-retest reliability protocol and 75 WYC (mean age 33 ± 5 yr) completed the validity protocol. Interclass correlation coefficients (ICC) between AWAS administrations and Spearman's correlation coefficients (rs) between AWAS and MTI data were calculated.

Results: AWAS showed good test-retest reliability (ICC = 0.80 (0.65-0.89)) and acceptable criterion validity (rs = 0.28, P = 0.01) for measuring weekly health-enhancing physical activity. AWAS also provided repeatable and valid estimates of sitting time (test-retest reliability, ICC = 0.42 (0.13-0.64); criterion validity, rs = 0.32, (P = 0.006)).

Conclusion: The measurement properties of the AWAS are comparable to those reported for existing self-report measures of physical activity. However, AWAS offers a more comprehensive and flexible alternative for accurately assessing different domains and intensities of activity relevant to WYC. Future research should investigate whether the AWAS is a suitable measure of intervention efficacy by examining its sensitivity to change.

1School of Psychology, The University of Queensland, Brisbane, Queensland, AUSTRALIA; and 2School of Public Health, Queensland University of Technology, Brisbane, Queensland, AUSTRALIA

Address for correspondence: Brianna S. Fjeldsoe, Cancer Prevention Research Centre, School of Psychology, The University of Queensland, Herston, Brisbane, Queensland, 4006, Australia; Email:

Submitted for publication October 2008.

Accepted for publication November 2008.

Despite the numerous health benefits associated with regular participation in physical activity (13), adult population participation levels remain at stable but unfavorably low levels in many industrialized countries (7,15,21,23,27,28,34). Data suggest that women with young children (WYC) are less likely to meet physical activity levels recommended for health benefit compared with women of similar age without children (10). Further, over one quarter of US women in 2004 and 13% of Australian women in 2000 reported no leisure-time physical activity (7,15).

There is evidence supporting a broad range of positive health outcomes associated with physical activity among women (10), including data that identify physical "inactivity" as the third large st modifiable risk factor for disease and injury prevention among Australian women (4). However, there have been suggestions that current self-report measures underestimate women's physical activity participation (1,10,19) and that further research is required to clarify the relationship between physical activity and health outcomes in this population. Paramount to this research is the need for a measure of physical activity that can accurately and reliably capture the wide range of health-enhancing physical activities engaged in by women (1).

There is mounting evidence that time spent in sedentary behaviors and, in particular, sitting time is related to negative health outcomes and biomarkers of disease risk among adults (9,24,26,30). There is currently a lack of evidence related to sitting among WYC in terms of health outcomes, descriptive patterns of behavior, and measurement. There have recently been calls to develop reliable and valid measures of nonoccupational sitting time (18).

The Australian Women's Activity Survey (AWAS) was developed after systematic and formative review of the capacity of existing measures of physical activity to accurately assess the wider range of health-enhancing activities emerging in the literature, including walking or cycling for transport, physically demanding occupational tasks (29), gardening tasks, and other labor-intensive domestic chores (6,11). The review also targeted measures of sitting behavior due to the recently identified independent negative associations between sitting and various health outcomes (9,24,26,30).

The Typical Week Physical Activity Survey (TWPAS) (2) was initially identified as one measure that may suit the needs identified above. However, formative review of the TWPAS with Australian WYC (19) revealed that women had difficulty conceptualizing their activity across the 11 categories defined in the TWPAS and that they needed more precise definition of activity intensity (including graphical representation). Women also requested that the instrument directly assess their sitting (or lack thereof) and low-intensity activities, provide multiple, relevant examples of activities for each activity category, be interview administered to allow for participant clarification, and assess activities separately for weekdays and weekend days (19). Developmental decisions for AWAS were also based on researchers' need to identify domain-specific activities being targeted in behavioral interventions. Therefore, AWAS was designed to specifically delineate activities that may be specifically promoted in physical activity intervention research (i.e., walking for exercise, walking for transport).

Like the TWPAS, AWAS uses a typical week in the past month as the reference period and asks respondents to recall the frequency (d·wk−1) and duration (mins/day) of a variety of activities during weekdays and weekend days. However, to meet the needs described above, AWAS separates activity into five domains (planned activities, employment, child care, domestic responsibilities, and transport) across five consistent intensity levels (sitting, light intensity, brisk walking, moderate intensity, and vigorous intensity). An activity category refers to a specific intensity level within a certain domain (i.e., light-intensity domestic responsibilities). Pilot testing of a self-complete version of AWAS revealed poor completion rate; therefore, AWAS is designed to be interviewer administered.

The aim of this study was to evaluate the measurement properties of the face-to-face, interview-administered AWAS among Australian WYC. Seven-day test-retest reliability and criterion validity against a Manufacturing Technology Inc. (MTI) accelerometer were assessed.

Back to Top | Article Outline


Back to Top | Article Outline

Sample recruitment.

Women were invited to participate through flyers and e-mails sent via a mother's playgroup association and were asked to contact research staff to register their interest in participating. Eligible subjects had at least one child aged less than 5 yr, were less than 14 wk pregnant, and were able to speak and read English. Because body proportions alter and fetal movement increases with gestation (32), subjects were required to be less than 14 wk pregnant to avoid possible accelerometer measurement errors due to the sensitivity of the monitor placement around the waist in later stages of pregnancy. To detect a hypothesized minimum correlation of 0.4 for criterion validity, which is viewed as an acceptable correlation for the validity of physical activity surveys (33,38), 47 subjects were required (assuming alpha = 0.05, power = 0.80). The study was approved by a university human research ethics committee.

Back to Top | Article Outline

Study design.

Subjects provided written informed consent to participate, completed the interview-administered AWAS on two consecutive visits (7 d apart), and wore an MTI accelerometer for the 7-d period between administrations. A demographic survey was completed at the first visit. Subjects were given a laminated instruction card (Appendix A) to refer to while answering questions during the face-to-face AWAS interview. The interviewer followed a script to ensure consistency between survey administrations (Appendix A). After completing the AWAS, each subject received individual instruction on how to wear the accelerometer and was asked to wear it around their waist for all waking hours for the next 7 d, commencing immediately. Subjects were also asked to remove the accelerometer anytime they came into contact with water (i.e., showering, swimming) and to record these times in a logbook.

Back to Top | Article Outline

Data preparation.

Estimates of total time spent in each activity category reported in AWAS were calculated by multiplying the reported minutes by reported frequency of the activity for weekdays and weekends separately. The sum of the weekday and weekend minutes was calculated to estimate total time spent in each activity category over a typical week. Total activity for each intensity level was calculated by summing the weekly minutes from each domain. Total activity for each domain was calculated by summing the weekly minutes from each intensity level within each domain. Total weekly health-enhancing physical activity (HEPA) was calculated by summing data from the activity domains that are widely accepted as sufficient to confer health benefit (i.e., brisk walking, moderate- and vigorous-intensity activities reported during leisure time and/or transport) (25). This HEPA total is consistent with recommendations for treating data collected using other existing self-report measures (i.e., Active Australia Questionnaire (AAQ) (3), Behavioral Risk Factor Surveillance System (BRFSS) physical activity module (14), International Physical Activity Questionnaire (IPAQ) [20]).

Accelerometer data were considered valid if there was more than 600 min of monitoring per day (excluding strings of zeros 20 min or longer) and four or less bouts of 20-min strings of zeros recorded per day (because these data strings suggest nonwear time). Subject's data were included in the analysis if they had at least three valid days of monitoring, one of which had to be a weekend day. This amount of accelerometer data is considered to be sufficient to determine habitual physical activity (37). The subjects did not systematically use the logbooks or record any substantial amounts of physical activity in them, so the logbook records were not considered when analyzing the accelerometer data.

Accelerometer counts were categorized as light, moderate, or vigorous intensity using two previously published cut points, one by Freedson et al. (22) and the other by Swartz et al. (35). The Freedson cut points are based on treadmill-based walking and running activities (22), whereas the Swartz cut points are based on field-based activities such as yard work, family care, and housework (35). To estimate time spent sitting, accelerometer counts between 0 and 100 were summed, excluding zero counts accumulated in strings of 20 min or more (20). The time recorded in each intensity range on each valid day was summed and divided by the number of valid days to provide an average day estimate. The average was then multiplied by seven to provide estimates of weekly minutes to allow for comparison to AWAS data.

Back to Top | Article Outline

Statistical analysis.

All analyses were performed using SPSS v16 and assumed an alpha level of 0.05. AWAS and accelerometer data were significantly skewed (according to the Kolmogorov-Smirnov tests); thus, medians and interquartile ranges (IQR) were used to describe the distributions of the data.

Interclass correlation coefficients (ICC) were calculated on log-transformed AWAS data using a two-way model to account for between administration variations and was reported for a single administration rather than the average of the two administrations because this is a more accurate reflection of intended use of the AWAS (31). Ninety-five percent confidence intervals were calculated for each ICC to demonstrate the precision of the reliability estimate, which is more accurate in reliability reporting than significance testing (31). An ICC of 0.80 or greater between physical activity survey administrations is considered ideal (5); however, lower correlation coefficients have been reported as acceptable (12,36). Test-retest reliability was assessed for all the intensity categories across all domains and also for all the domains across all the intensity categories.

To assess AWAS criterion validity, Spearman correlation coefficients (rs) were calculated to examine the relationship between weekly minutes of AWAS data (time 1) and activity time recorded by the accelerometers (using the Freedson and Swartz cut points). Validity was assessed for each intensity category across all domains (domain-specific criterion validity is unable to be established using accelerometers, which are unable to detect the context or type of physical activity). The validity of the HEPA estimate was also calculated to allow for comparison of validity outcomes to other questionnaires that only assess these domains of activity (brisk walking and moderate- and vigorous-intensity activities reported during leisure time and/or transport).

Back to Top | Article Outline


Forty subjects completed the reliability protocol, and 75 subjects completed the validity protocol. The larger validity protocol sample size was due to the need for an additional phase of data collection due to low rates of usable accelerometer data in the first phase. Overall, 53 subject's accelerometer data were not usable (due to inadequate wear time).

Table 1 shows a summary of demographic characteristics of the reliability and the validity samples. There was a significantly higher proportion of subjects with low education levels (year 10 or less) in the validity sample compared with the reliability sample (χ2 = 81.21, P < 0.00; see Table 1). There were no other statistically significant differences in demographic characteristics of subjects between the reliability and the validity samples (Table 1). There were also no statistically significant differences in demographic characteristics or AWAS-reported HEPA between subjects who met the accelerometer wear criteria (n = 75) and those who did not meet the criteria (n = 53).

Back to Top | Article Outline

AWAS test-retest reliability results

ICC for the test-retest reliability of the AWAS intensity categories ranged from 0.42 (sitting) to 0.80 (HEPA) (Table 2). ICC for the AWAS activity domains ranged from 0.62 (domestic activities) to 0.79 (planned activities) (Table 2). Sitting had a relatively low ICC between AWAS administrations (0.42 (95% CI = 0.13-0.64)). The median time reported sitting per week in the AWAS increased between administrations (Table 2).

Back to Top | Article Outline

AWAS criterion validity results.

Criterion validity correlations between the AWAS and the accelerometer data using Freedson and Swartz cut points are shown in Table 3. Using Freedson cut points, the correlation coefficients ranged from 0.07 (vigorous-intensity activity, all domains) to 0.36 (vigorous-intensity activity, planned and transport domains) (Table 3), whereas when using the Swartz cut points, the range of validity correlation coefficients was 0.05 (moderate-intensity activity, all domains) to 0.33 (vigorous-intensity activity, planned and transport domains).

There was a substantial difference in the validity correlations of the moderate-intensity and vigorous-intensity categories depending on which activity domains were included, particularly when using the Freedson cut points. The correlation between the AWAS and the accelerometer counts (using Freedson cut points) when all AWAS activity domains were included was 0.11 for moderate-intensity activities and 0.07 for vigorous-intensity activities. When only planned and transport activity domains were considered, the correlations strengthened to 0.22 for moderate-intensity activities and 0.36 for vigorous-intensity activities. Time reported sitting in the AWAS had a significant correlation with time recorded sitting by the accelerometer (rs = 0.32, P = 0.006).

Table 3 also shows the median time reported per week in the AWAS and recorded by the accelerometers. The medians reported in the AWAS were closer to the medians estimated from the Swartz cut points than they were using the Freedson cut points. This trend was particularly evident for moderate-intensity activity. The median minutes in moderate-intensity physical activity per week reported in the AWAS was 1320 (IQR = 990), and the estimate from accelerometers was 1070 (IQR = 541) when using Swartz cut points and 132 (IQR = 96) when using Freedson estimates. AWAS-reported time per week in moderate-intensity activities was substantially higher when all domains were included than moderate-intensity activity in planned and transport activity domains. This led to a stronger correlation of the AWAS-reported moderate-intensity planned and transport activity with the Freedson-derived accelerometer estimates for moderate-intensity activity (rs = 0.22) than with the Swartz-derived accelerometer estimates for moderate-intensity activity (rs = 0.06).

Back to Top | Article Outline


On the basis of the findings of this study, the AWAS provides reliable and valid estimates of planned moderate- to vigorous-intensity activities among Australian WYC. The ICC test-retest reliability coefficient for moderate-intensity physical activity in the AWAS was 0.74, which was higher than that reported for the Active Australia Questionnaire (AAQ) (ICC = 0.16 (12) or 0.52 (36)) but slightly lower than the ICC reported for the Pregnancy Physical Activity Questionnaire (PPAQ) (ICC = 0.82) (16). However, the PPAQ uses a categorical measure of physical activity, not a continuous measure like the AWAS; therefore, there is less variability in the PPAQ responses and thus stronger test-retest reliability. The criterion validity coefficient for moderate-intensity physical activity in the AWAS (using Freedson cut points, rs = 0.22) was similar to that reported for the PPAQ (rs = 0.20) (16), the AAQ (rs = 0.19) (36), and the BRFSS (r = 0.27) (39). The criterion validity coefficient for reported weekly sitting time in the AWAS (rs = 0.32) was the same as that reported for the IPAQ (rs = 0.32) (20). Therefore, the measurement properties of the AWAS are similar to those reported for other self-report measures of physical activity.

Although the AWAS may have similar measurement properties as existing phy sical activity surveys, it has some practical advantages over other self-report measures currently in use. The AWAS assesses a wide range of activity domains commonly engaged in by WYC, including activity domains not captured by existing surveys (i.e., child care activities, household responsibilities). The AWAS collects data as a continuous measure of minutes per week, unlike other surveys that extrapolate from categorical responses to obtain an estimate of weekly energy expenditure (16). Further, AWAS data enables analysts to look at intensity- and domain-specific activity (i.e., brisk walking for transport) for weekdays and weekend days separately if required. Such specificity enhances our capacity to answer research questions about the patterns and the properties of specific types and intensities of physical activity and to conduct more comprehensive and valid evaluation of interventions that target specific physical activity behaviors. This level of data collection also enables researchers to calculate estimates of energy expenditure (i.e., MET·min·wk−1), which allows them to explore the influence of energy expenditure from a wide range of activities on various health outcomes.

Researchers have recently become interested in the independent effects of sedentary behaviors on health outcomes (26). This growing body of research relies on accurate measures of domain specific sedentary behaviors and particularly sitting (18). The test-retest reliability of the AWAS sitting category (ICC = 0.42) was within the range of test-retest ICC of a recent review of sitting measures (18). However, it is difficult to compare the reliability results across surveys because most of the surveys assess one type of sitting behavior, such as sitting while watching television. Similar criterion validity correlations were observed for AWAS sitting data as those identified for sitting data collected by the IPAQ (20). However, the IPAQ measure only provides a generic estimate of sitting (20). The AWAS, with its domain-specific measure of sitting time, may be a useful tool for monitoring sitting time across weekdays and weekends, at least among WYC.

An interesting finding from this study was the variation in the criterion validity coefficients observed across the different AWAS activity domains and intensity categories. For example, validity coefficients for planned and transport activity domains were the strongest, but when data from the child care; and the domestic activity domains were included, the validity coefficients weakened. This phenomenon may have several explanations. It may be that waist-mounted accelerometers, such as the ones used in this study, are not designed to accurately capture movements typical of child care and household activity (e.g., upper body movements) (17). Activities associated with child care and domestic chores (such as carrying a child upstairs or scrubbing a bathtub) may not be accurately captured by the accelerometers, weakening the correlation between these activities as reported in the AWAS and as recorded by accelerometers.

Another possible explanation for differential correlations is that WYC find it difficult to accurately recall child care and domestic activities. In the formative research underpinning the development of AWAS (19), women reported it was more difficult to quantify and compartmentalize domestic and child care activities than planned or exercise activities because they happened sporadically throughout a typical day and not in defined time segments (19). It is therefore important to isolate these activities from the other domains of physical activity when attempting to capture what is currently accepted as HEPA (planned or leisure activity). A strength of the design of the AWAS is that analysts can choose which combination of data (intensity and/or domain) can be summed for analysis, which is important given that the definition of what should be included in HEPA is dependent on the health outcome being studied and its underlying biological mechanisms (20).

So far, only brisk walking, moderate- and vigorous-intensity leisure-time physical activity, and transport-related activities are typically included in the calculation of HEPA (1,33). Recently, a more encompassing view of activities of daily living has become the focus of physical activity measures (8). As a result, the total weekly physical activity estimates derived from measures such as the AWAS and the IPAQ will be higher than those recorded by other instruments, which do not measure such a wide range of activity domains. This will result in greater estimates of the proportion of people meeting the current physical activity guidelines (25) and thus the estimated prevalence of physical activity. Movement toward more inclusive assessment of physical activity has implications for the criteria and recommendations used to classify individuals as sufficiently or insufficiently active and may subsequently affect physical activity recommendations and intervention strategies. It is therefore crucial that further research examines the potential contribution of household, child care, and occupational activities to determine their health effects under varying conditions of participation (including length of each bout, varying intensities within each activity domain, etc.). The availability of a comprehensive yet flexible, valid, and reliable measure like AWAS can only assist in the endeavor to elucidate the health effects of different domains and intensities of activity such that more accurate recommendations can be made.

Another important finding in this study was the difference in validity correlations when accelerometer data were analyzed using the Freedson cut points (22) compared with the Swartz cut points (35). The validity correlation coefficients remained fairly stable when considering all activity domains, regardless of which cut point was used. However, the median minutes per week recorded by the accelerometers were much closer to the AWAS-reported estimates when using the Swartz cut points compared with the Freedson cut points. This suggests that the field-derived Swartz cut-point ranges are more closely aligned with what the women perceive to be light, moderate, or vigorous activity and what they report in each of these intensity categories in the AWAS. However, when only considering moderate-intensity activities from the planned and the transport activity domains, the validity correlation coefficient using the Freedson cut points was much stronger than when using the Swartz cut points. The majority of AWAS-reported moderate-intensity activity was reported in the domestic responsibilities or household activity domains and therefore excluded from this activity estimate. Therefore, the small amount of planned and transport-related moderate-intensity activity was weakly correlated to the larger estimate of Swartz-derived accelerometer data. This problem is difficult to overcome because the accelerometers cannot collect domain-specific information.

As is common in assessing the test-retest reliability of any self-report measure, the AWAS data collected at time 2 may have been affected by the subject's recall at time 1. Further, subject's recall of their physical activity behavior at time 2 may have been influenced by wearing the accelerometers in the preceding week. For example, subjects may have increased their physical activity while wearing the accelerometers to demonstrate socially desirable behaviors while being monitored. Furthermore, subjects may have been more aware of their daily activities at time 2 if wearing an accelerometer or recently completing the AWAS at time 1 raised consciousness about their physical activity participation. These methodological issues are present in any study using the same study design to test the measurement properties of a self-report measure of physical activity (12,16,20,36) and are difficult to circumvent.

A methodological issue raised in this study was the large proportion of subject's data that did not meet the accelerometer wear criteria. Forty-one percent of subjects did not wear the accelerometer for a minimum of 10 h·d−1 for 3 d (one being a weekend day). This high rate of data loss should be considered when planning to collect accelerometer data in this population in future research. Incentives for wearing accelerometers may be required. There were no differences in demographic or self-reported physical activity between subjects who met the accelerometer wear criteria and those who did not; therefore, the high rate of missing data should not have affected the validity correlations of the AWAS.

This study has shown that AWAS data provide reliable and valid estimates of HEPA among Australian WYC. The AWAS can be considered a valuable addition to the existing portfolio of self-report physical activity surveys because it assesses a wide range of activities, collects data in a format that allows researchers flexibility in analysis, and assesses sitting as an independent sedentary behavior. AWAS was administered via face-to-face interview; however, telephone administration may be a viable alternative after further investigation. Further exploration of how AWAS may be used in intervention research, and its sensitivity for detecting meaningful change in physical activity behavior is required.

The authors thank Dr. Stewart Trost for advice and support in early attempts to analyze the accelerometer data. Accelerometer data were finally analyzed using a program developed by Paul Jackson at the School of Psychology at The University of Queensland. Thank you to the Playgroup Association of Queensland for allowing us into their mothers' groups. Finally, thank you to Prof. Barbara Ainsworth for sharing with us the original version of the TWPAS that was used in the development phases of this work. This research was funded by a grant awarded to Dr. Yvette Miller from The University of Queensland New Staff Research Start-Up Fund. The results of the present study do not constitute endorsement by the American College of Sports Medicine.

Back to Top | Article Outline


1. Ainsworth BE. Issues in the assessment of physical activity in women. Res Q Exerc Sport. 2000;71:37-42.
2. Ainsworth BE, Lamonte MJ, Drowatzky KL. Evaluation of the CAPS Typical Week Physical Activity Survey among Minority Women. In: Proceedings of the Community Prevention Research in Women's Health Conference. Bethesda (MD): National Institute of Health, 2000. p. 17.
3. Armstrong T, Bauman A, Davies J. Physical Activity Patterns of Australian Adults. Results of the 1999 National Physical Activity Survey. Canberra: Australian Institute of Health and Welfare; 2000. p. 16-8.
4. Australian Institute of Health and Welfare. Australia's Health 2006. Canberra: Australian Institute of Health and Welfare; 2006. p. 162-5.
5. Baranowski T, Moor C. How many days was that? Intra-individual variability and physical activity assessment. Res Q Exerc Sport. 2000;71:74-8.
6. Bassett D, Ainsworth B, Swartz A, Strath S, O'Brien W, King A. Validity of four motion sensors in measuring moderate-intensity physical activity. Med Sci Sports Exerc. 2000;32(9 suppl):S471-80.
7. Bauman A, Ford I, Armstrong T. Trends in Population Levels of Reported Physical Activity in Australia, 1997, 1999 and 2000. Canberra: Australian Sports Commission; 2001. p. 4-27.
8. Bauman A, Allman-Farinelli M, Huxley R, James W. Leisure-time physical activity alone may not be a sufficient public health approach to prevent obesity-a focus on China. Obes Rev. 2008;1:119-26.
9. Blanck H, McCullough M, Patel A, et al. Sedentary behavior, recreational physical activity, and 7-year weight gain among post-menopausal US women. Obesity. 2007;15:1578-88.
10. Brown W, Mishra G, Lee C, Bauman A. Leisure time physical activity in Australian women: relationship with well-being and symptoms. Res Q Exerc Sport. 2000;71:206-16.
11. Brown W, Ringuet C, Trost S, Jenkins D. Measurement of energy expenditure of daily tasks among mothers of young children. J Sci Med Sport. 2001;4:379-85.
12. Brown W, Trost S, Bauman A, Mummery Owen N. Test-retest reliability of four physical activity measures used in population surveys. J Sci Med in Sport. 2004;7:205-15.
13. Bull FC, Bauman AE, Bellew B, Brown W. Getting Australia Active II: An Update of Evidence on Physical Activity and Health. Melbourne (Australia): National Public Health Partnership; 2004. p. 7-42.
14. Centre for Disease Control and Prevention. 2003 Behavioral Risk Factor Surveillance System State Questionnaire, Volume 15. Atlanta (GA): US Department of Health and Human Services, Centers for Disease Control and Prevention; 2002. p. 1-5.
15. Centre for Disease Control and Prevention. Trends in leisure-time physical inactivity by age, sex and race/ethnicity-United States, 1994-2004. Morb Mortal Wkly Rep. 2005;54:991-4.
16. Chasan-Taber L, Schmidt M, Roberts D, Hosmer D, Markenson G, Freedson P. Development and validation of a pregnancy physical activity questionnaire. Med Sci Sports Exerc. 2004;36(10):1750-60.
17. Chen K, Bassett D. The technology of accelerometry-based activity monitors: current and future. Med Sci Sports Exerc. 2005;37(11 suppl):S490-500.
18. Clark BK, Sugiyama T, Healy G, Salmon J, Dunstan D, Owen N. Validity and reliability of measures of television viewing time and other non-occupational sedentary behavior of adults: a review. Obes Rev. 2008;1-10.
19. Collins BS, Marshall AL, Miller YD. Physical activity in women with young children: how can we assess 'anything that's not sitting'? Women Health. 2007;45:95-116.
20. Craig C, Russel S, Cameron C, Bauman A. Twenty-year trends in physical activity among Canadian adults. Can J Public Health. 2004;95:59-63.
21. Craig CL, Marshall AL, Sjöström M, et al. International Physical Actitvity Questionaire: 12 - country reliability and validity. Med Sci Sports Exerc. 2003;35(8):1381-95.
22. Freedson P, Melanson E, Sirard J. Calibration of the Computer Science and Applications Inc. accelerometer. Med Sci Sports Exerc. 1998;30(5):777-81.
23. Haase A, Steptoe A, Sallis J, Wardle J. Leisure-time physical activity in university students from 23 countries: associations with health beliefs, risk awareness and national economic development. Prev Med. 2004;39:182-90.
24. Hamilton M, Hamilton D, Zderic T. Role of low energy expenditure and sitting in obesity, metabolic syndrome, type 2 diabetes and cardiovascular disease. Diabetes. 2007;56:2655-67.
25. Haskell W, Lee I, Pate R, et al. Physical activity and public health: updated recommendation for adults from the American College of Sports Medicine and the American Heart Foundation. Med Sci Sports Exerc. 2007;39(8):1423-34.
26. Healy GJ, Salmon J, Wijndaele K, et al. Objectively measured sedentary time, physical activity, and metabolic risk. Diabetes Care. 2008;31:369-71.
27. Hillsdon M, Cavil N, Nanchahal K, Diamont A, White I. National level promotion of physical activity: results from England's ACTIVE for LIVE campaign. J Epidemiol Community Health. 2000;55:755-61.
28. Macera CA, Ham SA, Yore MM, et al. Prevalence of physical activity in the United Sates: Behavioral Risk Factor Surveillance System, 2001. Prev Chronic Dis. 2005;2:1-10.
29. Pate R, Pratt M, Blair S, Haskell W, Macera C, Bouchard C. Physical activity and public health: a recommendation from the Centres of Disease Control and Prevention and the American College of Sports Medicine. J Am Medical Assoc. 1995;273:402-7.
30. Patel A, Rodriguez C, Pavluck A, Thun M, Calle E. Recreational physical activity and sedentary behavior in relation to ovarian cancer in a large cohort of US women. Am J Epidemiol. 2006;163:709-16.
31. Patterson P. Reliability, validity, and methodological responses to the assessment of physical activity via self-report. Res Q Exerc Sport. 2000;71:15-20.
32. Rousham EK, Clarke PE, Gross H. Significant changes in physical activity among pregnant women in the UK as assessed by accelerometry and self-reported activity. Eur J Clin Nutr. 2006;60:393-400.
33. Sallis JF, Saelens BE. Assessment of physical activity by self-report: status, limitations, and future directions. Res Q Exerc Sport. 2000;71:1-14.
34. Steffen LM, Arnett DK, Blackburn H, et al. Population trends in leisure-time physical activity: Minnesota Heart Survey, 1980-2000. Med Sci Sports Exerc. 2006;38(10):1716-23.
35. Swartz AM, Strath SJ, Bassett DR, O'Brien WL, King GA, Ainsworth BE. Estimation of energy expenditure using CSA accelerometers at hip and wrist sites. Med Sci Sports Exerc. 2000;32(9 suppl):S450-6.
36. Timperio A, Salmon J, Crawford D. Validity and reliability of a physical activity recall instrument among overweight and non-overweight men and women. J Sci Med Sport. 2003;6:477-91.
37. Trost SG, McIver KL, Pate RR. Conducting accelerometer-based activity assessments in field-based research. Med Sci Sports Exerc. 2005;37(11 suppl):S531-43.
38. Washburn R, Heath G, Jackson A. Reliability and validity issues concerning large scale surveillance of physical activity. Res Q Exerc Sport. 2000;71:104-13.
39. Yore MM, Ham SA, Ainsworth BE, et al. Reliability and validity of the instrument used in BRFSS to assess physical activity. Med Sci Sports Exerc. 2007;39(8):1267-74.
Back to Top | Article Outline

Australian Women's Activity Survey

Interview Script



©2009The American College of Sports Medicine